Cohorts and trials point the same way. The exact amount and the size of the benefit are less certain.
The claim
Eating more fiber lowers your risk of dying early and of heart disease.
What the evidence says
A series of reviews pooled 185 prospective studies and 58 clinical trials (4,635 people in the trials) in adults without chronic disease; trials with fiber supplements were excluded. Comparing the highest with the lowest fiber consumers, the observational data suggest about 15–30% lower risk of dying (all-cause and cardiovascular), coronary heart disease, stroke, type 2 diabetes and colorectal cancer. In trials, higher fiber intakes went with lower body weight, systolic blood pressure and total cholesterol. In the cohort data, the risk reduction was greatest at 25–29 g a day, and higher intakes may help even more. Certainty of evidence for fiber was graded moderate.
The numbers
- Reynolds 2019: 185 prospective studies and 58 clinical trials (4,635 trial participants), adults without chronic disease, supplement trials excluded.
- Observational: highest vs lowest fiber intake went with about 15–30% lower risk of all-cause and cardiovascular death, CHD, stroke, type 2 diabetes and colorectal cancer.
- Trials: lower body weight, systolic blood pressure and total cholesterol. Cohort data: risk reduction greatest at 25–29 g a day; higher intakes may help even more. GRADE certainty for fiber: moderate.
Verdict: Mostly proven
Cohorts and trials point the same way. The exact amount and the size of the benefit are less certain.
Why only this far: The mortality evidence is observational (relative differences between the highest and lowest groups); the trials measured risk factors, not deaths. GRADE certainty for fiber: moderate.
Evidence level B: Large cohort studies. How verdicts work.
What we don’t know
Cohort studies can’t rule out that fiber eaters live healthier in other ways. The trials measured weight, blood pressure and cholesterol, not deaths.
Sources
Checked 15 Oct 2026. Every link opens the original paper or its PubMed record.
- Reynolds et al., Lancet 2019Series of systematic reviews (185 prospective studies, 58 randomized trials)DOI 10.1016/S0140-6736(18)31809-9 · PubMed 30638909The Lancet later published a correction notice (“Department of Error”, Lancet 2019;393:406). The figures on this page follow the paper’s abstract.
Follow for the reel on Instagram
Educational content, not medical advice. This check rates a claim, not you or your health. Talk to a doctor about your own situation. Found a mistake? Write to hello@mostlyproven.com.
